Approximately 25% of people age 65 and older fall each year. These falls are a principal cause of injuries, death, and disability among the elderly. Falls occur more often among frail people, but are also common among the healthy. The injuries and fear of falling that often develop after a fall can increase the level of frailty and dependence of the elderly. Fractures are the most common serious injury from falls; as a result, falls are an important risk factor for some types of fractures. Hip fractures are the most costly and disabling type of fracture among the elderly, and are almost always caused by a fall. Increasing hip fracture incidence rates have been observed in successive generations of the same race, suggesting that environmental/lifestyle factors may influence the risk of fracture. Hip fracture incidence rates among certain minorities (Japanese-Americans, Blacks, and Mexican-Americans) are lower than among Caucasians. The rates of other non-spine fractures among Japanese-Americans is only one-third that reported for U.S. Caucasians. Since the bone mass of Japanese-Americans is lower than Caucasians, the lower incidence of fractures among Japanese may be related to other factors, such as the frequency of falls. In support of this hypothesis, we have found that elderly Japanese-American women in Hawaii fall fewer times per year on average than reported for U.S. Caucasians. We propose here to extend an existing longitudinal study of bone loss and fractures, and begin a prospective study of falls and fall-related injuries (including fractures) among elderly Japanese-American women in Hawaii. This study will explore whether or not differences in fracture incidence (and possibly other injuries) between Caucasians and Japanese are related to differences in the patterns of falls, bone mass, and other factors (such as muscle strength or body size). It will also explore what factors are responsible for the greater frequency of falls among Caucasians, compared to Japanese. For example, we will examine whether or not the observed difference in the rate of falls (or fractures, or other injuries) might be due to a difference in 1) the prevalence of risk factors, and/or 2) a difference in the strength of association with risk for falling for fractures or other injuries), between races. If the observed differences between races are related to potentially modifiable risk factors, rather than genetic differences, the results may help to design prevention strategies to reduce the frequency of falls and fractures among the elderly.